An L<sub>1</sub>-and-L<sub>2</sub>-Norm-Oriented Latent Factor Model for Recommender Systems.
basic_science · Level V
Where this comes from
- Record sourced from PubMed, PMID 33886475.
- Also identified by DOI 10.1109/TNNLS.2021.3071392.
- No licence information is recorded for this record.
- Because redistribution is not established, this page shows the abstract only. Follow the links below for the full text.
Abstract
A recommender system (RS) is highly efficient in filtering people's desired information from high-dimensional and sparse (HiDS) data. To date, a latent factor (LF)-based approach becomes highly popular when implementing a RS. However, current LF models mostly adopt single distance-oriented Loss like an L<sub>2</sub> norm-oriented one, which ignores target data's characteristics described by other metrics like an L<sub>1</sub> norm-oriented one. To investigate this issue, this article proposes an L<sub>1</sub> -and- L<sub>2</sub> -norm-oriented LF ( [Formula: see text]) model. It adopts twofold ideas: 1) aggregating L<sub>1</sub> norm's robustness and L<sub>2</sub> norm's stability to form its Loss and 2) adaptively adjusting weights of L<sub>1</sub> and L<sub>2</sub> norms in its Loss. By doing so, it achieves fine aggregation effects with L<sub>1</sub> norm-oriented Loss 's robustness and L<sub>2</sub> norm-oriented Loss 's stability to precisely describe HiDS data with outliers. Experimental results on nine HiDS datasets generated by real systems show that an [Formula: see text] model significantly outperforms state-of-the-art models in prediction accuracy for missing data of an HiDS dataset. Its computational efficiency is also comparable with the most efficient LF models. Hence, it has good potential for addressing HiDS data from real applications.